Senior Systems Architect

Hi, Jed here.
I like to build things.

Crafting high-performance distributed systems and sophisticated user interfaces at the intersection of design and engineering.

expand_more
/

Experience

2024 — PRESENT

Lead Ecommerce Solutions Engineer — Leanin' Tree Inc.

Lead the design and implementation of a multi-platform ecommerce ecosystem, specializing in Shopify development, automated ERP integration, and high-impact cost-saving migrations for diverse stakeholders.

TypeScriptGoShopify LiquidPythonReact Router / Remix

2021 — 2023

Senior Software Engineer — Pink Summit LLC

Split across two engineering teams - Spearheaded migration of a legacy Java Applet platform to Angular, modernizing user experiences, and built Python-based Atlassian integrations under government security requirements, handling sensitive data workflows in compliance-controlled AWS environments.

AngularPythonJavaTypeScriptAWS

2018 — 2021

Software Engineer — RxD Software LLC

Built bespoke web applications and backend services for a diverse client portfolio. Shipped production systems utilizing a wide range of technologies to meet unique project requirements.

PythonGoTypeScriptReactAWS

2014 — 2018

B.A. Computer Science — University of Colorado Boulder

Minor in Leadership. Studied software engineering, algorithms, and systems design while developing foundational skills in cross-functional collaboration and technical leadership.

Design Database SystemsHuman Computer InteractionIntro to AIAlgorithms
/

Personal Projects

code GitHub
candlestick_chart

AlgoTrader

GoPythonAWSTerraformLightGBMPostgreSQL

A fully autonomous algorithmic trading platform designed to execute complex equity strategies without human intervention. The system identifies high-potential stocks daily, analyzes market sentiment, and manages live capital with professional-grade risk controls to ensure consistent performance during market hours.

Challenge

Building a zero-intervention trading system that manages real capital across three float tiers, with robust risk controls, 4x PDT margin utilization, and reliable EOD liquidation, all on a cost-efficient serverless infrastructure.

Solution

Go trading engine on ECS Fargate with EventBridge scheduling (~81% cost reduction vs always-on), a 14-step risk approval pipeline, SQS FIFO for per-symbol ordering, weekly walk-forward ML retraining with SHAP pruning and KS drift detection, and a real-time Svelte dashboard via WebSocket push.

Technical Strategy

Go was chosen for the trading engine for its goroutine concurrency model and single-binary deployment, critical for processing 8 parallel symbol streams with sub-millisecond latency. Python handles the ML layer (LightGBM, FinBERT) where ecosystem maturity matters more than speed. ECS Fargate with EventBridge scheduling eliminates idle compute costs, while SQS FIFO guarantees per-symbol message ordering without a self-managed broker. Terraform codifies the entire stack for reproducible deploys.

open_in_new Live Site
fingerprint

SpectraSeal

RustFFmpegSIMDDCT/FFT

An advanced digital security ecosystem that protects media assets using invisible forensic watermarking. By embedding traceable data directly into the fabric of images and video, it allows companies to identify the source of unauthorized leaks even if the content has been screenshotted, compressed, or re-recorded.

Challenge

Traditional watermarks (visible overlays, LSB pixel-flipping) are trivially defeated by cropping, JPEG re-compression, or simply screenshotting the content, creating a new file with entirely different pixel values and potential geometric distortion.

Solution

Embeds AES-256-GCM encrypted payloads into mid-frequency DCT coefficients across 8x8 blocks, paired with an FFT-based pilot signal for geometric recovery. Video uses A/B adaptive streaming, dual-encoded segments form a per-user binary "DNA" sequence via manifest orchestration, eliminating per-user transcoding at scale.

Technical Strategy

Rust was chosen for its zero-cost abstractions and SIMD intrinsics (AVX2/NEON), critical for processing frequency blocks in parallel at <15ms image latency and real-time 60fps video. FFmpeg bindings handle segment encoding without reinventing codec logic. Reed-Solomon ECC and majority voting across blocks ensure blind extraction survives 50% JPEG quality, 720p→480p downscaling, and 15° rotation. The zero-copy ownership model eliminates unnecessary allocations in the media buffer pipeline.

open_in_new Live Site
palette

Chassis UI

LitWeb ComponentsStyle DictionaryAstroTurborepoTypeScript

A versatile, framework-agnostic UI library that eliminates 'vendor lock-in' for design systems. It allows engineering teams to share a single library of 32+ components across different web frameworks—like React, Svelte, and Angular—while instantly switching between 12 distinct visual identities (such as Shopify Polaris or Material Design) via simple CSS tokens.

Challenge

Design systems usually couple visual identity to specific framework code, forcing teams to rewrite their entire component library if they want to change their look or switch from React to Svelte. Additionally, preventing CSS leakage in complex enterprise apps is difficult without breaking customization capabilities.

Solution

Built a 'Design System Bridge' using Lit-based Web Components that leverage Shadow DOM for perfect encapsulation. The system features a library of 32 components across 6 categories, 12 pre-built themes, and a double-fallback token pattern. It includes auto-generated framework wrappers for React, Angular, Svelte, and Astro, ensuring a 'write once, use everywhere' workflow.

Technical Strategy

Lit was chosen for its minimal runtime footprint and native Web Component support, avoiding the overhead of a virtual DOM. Style Dictionary serves as the single source of truth, transforming JSON design tokens into CSS Custom Properties for sub-100ms theme switching. The architecture uses a monorepo with Turborepo and npm workspaces to orchestrate a complex build pipeline: processing tokens and themes first, then building the core components, and finally generating type-safe framework wrappers.

/

Expertise

Modern Stack 2026

Languages

TypeScript

Full-stack Development

Go (Golang)

Concurrency & Performance

Python

ML & Automation

Rust

Systems Programming

SASS

Styling & Design Systems

Bash

Scripting & Tooling

Frameworks & Runtime

React

Component Architecture

Angular

Enterprise Apps

Astro

Static & Hybrid Sites

Remix

Full-stack Routing

Node.js

Scalable I/O

Flask

Python Web Services

Data & Machine Learning

SQL

Postgres / Relational DBs

NoSQL

MongoDB / Document Stores

TensorFlow

Deep Learning

PyTorch

Neural Networks

Pandas

Data Analysis

Prompt Engineering

LLM Integration

Infrastructure & DevOps

AWS

Cloud Infrastructure

Docker

Containerization

Terraform

Infrastructure as Code

Firebase

Serverless & Realtime

Redis

Caching & Queues

GitHub Actions

CI/CD Pipelines

Tools & Integrations

GraphQLREST APIsWebhooksOAuthGitLiquidMaterialUIAtlassian PlatformJupyterShopifyFigmaStripeStrapiTailwind